Systems and methods for predicting destination in online to offline service

Abstract

A method may include determining that a service requester intends to request a service from a first location at a first time point. The method may include obtaining historical orders requested by the service requester in a first time period prior to the first time point. The method may include determining one or more candidate destinations based on the historical orders. The method may include, for each of the one or more candidate destinations, selecting at least one historical order based on the first time point. The method may include determining a probability associated with the each of the one or more candidate destinations based on a number of the selected at least one historical order and a number of historical orders associated with the each of the one or more candidate destinations. The method may include determining a destination based on the one or more probabilities.

1 . A system for predicting a destination in an online to offline service system, the system comprising:
 one or more storage devices storing a set of instructions; and   one or more processors configured to communicate with the one or more storage devices, wherein when executing the set of instructions, the one or more processors are configured to cause the system to:   determine that a service requester intends to request a service from a first location at a first time point;   obtain a plurality of historical orders requested by the service requester in a first time period prior to the first time point in response to the determination;   determine one or more candidate destinations based on the plurality of historical orders;   for each of the one or more candidate destinations,   select, from the plurality of historical orders, at least one historical order associated with the each of the one or more candidate destinations based on the first time point; and   determine a probability associated with the each of the one or more candidate destinations based on a number of the selected at least one historical order and a number of historical orders associated with the each of the one or more candidate destinations, wherein the probability indicates a likelihood that the service requester intends to travel to the each of the one or more candidate destinations at the first time point; and   determine, from the one or more candidate destinations, a destination to be recommended to the service requester based on the one or more probabilities associated with the one or more candidate destinations, respectively.   
     
     
         2 . The system of  claim 1 , the one or more processors are further configured to cause the system to:
 transmit the destination to a requester terminal associated with the service requester causing the destination displayed on a user interface of the requestor terminal.   
     
     
         3 . The system of  claim 1 , wherein a destination of the selected at least one historical order matches the each of the one or more candidate destinations, and a departure time associated with the selected at least one historical order is within a second time period that includes the first time point. 
     
     
         4 . The system of  claim 1 , wherein a departure location associated with the selected at least one historical order is within a distance range including the first location. 
     
     
         5 . The system of  claim 1 , wherein the number of the selected at least one historical order is determined by:
 determining a weight for each of the selected at least one historical order based on an interval between a departure time associated with the each of the selected at least one historical order and the first time point; and   determining the number of the selected at least one historical order based on a sum of the weights associated with the selected at least one historical order, respectively.   
     
     
         6 . The system of  claim 5 , wherein to determine the weight for each of the selected at least one historical order, the one or more processors are configured to cause the system to:
 determine a half-life based on the plurality of historical orders; and   determine the weight for the each of the selected at least one historical order based on the half-life and the interval between the departure time associated with each of the selected at least one historical order and the first time point.   
     
     
         7 . The system of  claim 1 , wherein to determine, from the one or more candidate destinations, the destination to be recommended to the service requester based on the one or more probabilities associated with the one or more candidate destinations, respectively, the one or more processors are configured to cause the system to:
 select a candidate destination with a maximum probability from the one or more candidate destinations;   determine whether the maximum probability exceeds a probability threshold; and   determine the candidate destination with the maximum probability as the destination to be recommended to the service requester in response to a determination that the maximum probability exceeds a probability threshold.   
     
     
         8 . A method for predicting a destination in an online to offline service system implemented on a computing device having one or more storage devices and one or more processors, the method comprising:
 determining that a service requester intends to request a service from a first location at a first time point;   obtaining a plurality of historical orders requested by the service requester in a first time period prior to the first time point in response to the determination;   determining one or more candidate destinations based on the plurality of historical orders;   for each of the one or more candidate destinations,   selecting, from the plurality of historical orders, at least one historical order associated with the each of the one or more candidate destinations based on the first time point; and   determining a probability associated with the each of the one or more candidate destinations based on a number of the selected at least one historical order and a number of historical orders associated with the each of the one or more candidate destinations, wherein the probability indicates a likelihood that the service requester intends to travel to the each of the one or more candidate destinations at the first time point; and   determining, from the one or more candidate destinations, a destination to be recommended to the service requester based on the one or more probabilities associated with the one or more candidate destinations, respectively.   
     
     
         9 . The method of  claim 8 , the method further comprising:
 transmitting the destination to a requester terminal associated with the service requester causing the destination displayed on a user interface of the requestor terminal.   
     
     
         10 . The method of  claim 8 , wherein a destination of the selected at least one historical order matches the each of the one or more candidate destinations, and a departure time associated with the selected at least one historical order is within a second time period that includes the first time point. 
     
     
         11 . The method of  claim 8 , wherein a departure location associated with the selected at least one historical order is within a distance range including the first location. 
     
     
         12 . The method of  claim 8 , wherein the number of the selected at least one historical order is determined by:
 determining a weight for each of the selected at least one historical order based on an interval between a departure time associated with the each of the selected at least one historical order and the first time point; and   determining the number of the selected at least one historical order based on a sum of the weights associated with the selected at least one historical order, respectively.   
     
     
         13 . The method of  claim 12 , wherein the determining of the weight for each of the selected at least one historical order includes:
 determining a half-life based on the plurality of historical orders; and   determining the weight for the each of the selected at least one historical order based on the half-life and the interval between the departure time associated with each of the selected at least one historical order and the first time point.   
     
     
         14 . The method of  claim 8 , wherein the determining of the destination to be recommended to the service requester based on the one or more probabilities associated with the one or more candidate destinations from the one or more candidate destinations includes:
 selecting a candidate destination with a maximum probability from the one or more candidate destinations;   determining whether the maximum probability exceeds a probability threshold; and   determining the candidate destination with the maximum probability as the destination to be recommended to the service requester in response to a determination that the maximum probability exceeds a probability threshold.   
     
     
         15 - 21 . (canceled) 
     
     
         22 . A non-transitory computer readable medium, comprising at least one set of instructions for predicting a destination in an online to offline service system, wherein when executed by one or more processors of a computing device, the at least one set of instructions causes the computing device to perform a method, the method comprising:
 determining that a service requester intends to request a service from a first location at a first time point;   obtaining a plurality of historical orders requested by the service requester in a first time period prior to the first time point in response to the determination;   determining one or more candidate destinations based on the plurality of historical orders;   for each of the one or more candidate destinations,   selecting, from the plurality of historical orders, at least one historical order associated with the each of the one or more candidate destinations based on the first time point; and   determining a probability associated with the each of the one or more candidate destinations based on a number of the selected at least one historical order and a number of historical orders associated with the each of the one or more candidate destinations, wherein the probability indicates a likelihood that the service requester intends to travel to the each of the one or more candidate destinations at the first time point; and   determining, from the one or more candidate destinations, a destination to be recommended to the service requester based on the one or more probabilities associated with the one or more candidate destinations, respectively.   
     
     
         23 . The non-transitory computer readable medium of  claim 22 , the method further comprising:
 transmitting the destination to a requester terminal associated with the service requester causing the destination displayed on a user interface of the requestor terminal.   
     
     
         24 . The non-transitory computer readable medium of  claim 22 , wherein a destination of the selected at least one historical order matches the each of the one or more candidate destinations, and a departure time associated with the selected at least one historical order is within a second time period that includes the first time point. 
     
     
         25 . (canceled) 
     
     
         26 . The non-transitory computer readable medium of  claim 22 , wherein the number of the selected at least one historical order is determined by:
 determining a weight for each of the selected at least one historical order based on an interval between a departure time associated with the each of the selected at least one historical order and the first time point; and   determining the number of the selected at least one historical order based on a sum of the weights associated with the selected at least one historical order, respectively.   
     
     
         27 . The non-transitory computer readable medium of  claim 26 , wherein the determining of the weight for each of the selected at least one historical order includes:
 determining a half-life based on the plurality of historical orders; and   determining the weight for the each of the selected at least one historical order based on the half-life and the interval between the departure time associated with each of the selected at least one historical order and the first time point.   
     
     
         28 . The non-transitory computer readable medium of  claim 22 , wherein the determining of the destination to be recommended to the service requester based on the one or more probabilities associated with the one or more candidate destinations from the one or more candidate destinations includes:
 selecting a candidate destination with a maximum probability from the one or more candidate destinations;   determining whether the maximum probability exceeds a probability threshold; and   determining the candidate destination with the maximum probability as the destination to be recommended to the service requester in response to a determination that the maximum probability exceeds a probability threshold.